A Practical Guide to Backing Up a Live Signal Path

Start by drawing the whole path on one page

Before anyone talks about equipment, I ask for a single sheet that shows the signal from the camera at the venue to the receive point at the broadcaster. Every box on that sheet is a place where the feed can fail, and every line between boxes is a cable, a link or a handover between two teams. Producers are often surprised by how many handovers there are.

Mark who owns each stage. A venue may supply the power and the building's network, a contractor may supply the uplink, and the broadcaster may supply the return feed. When a picture drops during a mid autumn evening broadcast, the first question is always whose stage it was. A sheet with names and phone numbers beside each box answers it in seconds.

Keep the drawing plain. Boxes, arrows, labels, and a note of the format passing through each link. Once it exists, the backup conversation becomes concrete rather than vague, because you can point at the exact box you are worried about and ask what happens if it goes dark.

• One page, camera to receive point
• Name the owner of every stage
• Note the video and audio format on each link

Decide which failures you can live with

Not every box needs a spare. A backup route costs money, crew time and rack space, so the useful question is which failures would take the programme off air and which would only cost you a camera angle. Losing one of four cameras is an inconvenience. Losing the single uplink from the venue is the end of the broadcast.

I sort the boxes into three groups: those that must never fail, those that can fail if there is a quick workaround, and those that simply do not matter on the night. Spend the backup budget on the first group. The second group usually needs a spare cable, a spare unit or a clear instruction to the operator, not a duplicate system.

Be honest with the producer about the sorting. A client who understands that the main link has a second route, while a single lock-off camera does not, will make better decisions about the running order. It also prevents the awkward conversation afterwards about why something unprotected was unprotected.

Build the second route so it does not share the first

A backup that shares a weak point with the main route is not a backup. If both feeds leave the building through the same cabinet, or both depend on the same power circuit, one fault removes both. The most common mistake I see is a second link that looks separate on the drawing but runs through the same physical cable run.

Where possible, use a different path in kind as well as in location. A fibre or dedicated circuit as the main route, with a bonded connection or a separate carrier as the standby, means one outage does not remove both. Power the two routes from separate circuits, and ask the venue where each one is fed from rather than assuming.

Then decide how the switch happens. Some changeovers are automatic, some need a person at a panel. Neither is wrong, but the team must know which one they have. An automatic switch nobody has tested is as risky as a manual one nobody is watching.

• Different carrier or medium where you can
• Separate power circuits for each route
• Know whether the changeover is automatic or manual

Put monitoring where you can actually see it

Monitoring is what turns a backup from a hope into a plan. You need to see the signal at more than one point: leaving the venue, arriving at the broadcaster, and ideally somewhere in between. If you only watch the final picture, you learn about a fault when the viewer does, and you have no idea which stage caused it.

Give someone the job of watching. A wall of small monitors is useless if the person responsible is also operating a camera or answering the producer. On our outside broadcasts, one engineer owns the signal view and nothing else, and they have the authority to call a switch to the standby route without asking permission.

Agree the warning signs beforehand. Frozen frames, audio dropping out, a creeping delay between the two ends, errors on a link. When everyone has seen the same fault example in rehearsal, the call to switch is made quickly and without argument.

Rehearse the failure, not just the show

A rehearsal that only proves the picture works proves very little. The valuable test is pulling the main link on purpose, with the broadcaster watching, and seeing how long it takes to recover. You learn how many seconds the gap lasts, whether audio returns with the picture, and whether the receiving end needs a nudge to lock back on.

Do this early enough to fix what you find. Arriving on site in the afternoon and testing an hour before air leaves no room to change anything. A day before, or at least the morning, gives you time to swap a cable, change a setting or ring the carrier.

Write down the result in plain words and share it with the producer and the receiving desk. If the gap is eight seconds, say eight seconds. People plan better around a known gap than around a reassuring promise that it will be seamless.

• Pull the main link deliberately
• Time the gap and note what returns first
• Test with the receiving end watching

Keep a short log through the broadcast

During the broadcast, keep a simple running log: the time of any change, what was seen, what was done and who was told. It does not need to be elaborate. A page of notes written as things happen is worth far more than anyone's memory the next morning, when the details have blurred.

The log serves two purposes. In the moment, it stops two people fixing the same problem in different ways. Afterwards, it tells you whether a fault was a one-off or a pattern, which decides whether you change the route next time or simply accept the risk.

Share the log with the client once the programme is finished. It shows what was protected, what was used and what was learned, and it feeds the next planning sheet. Over a season of broadcasts, those sheets become the most useful thing we own.